search

agentos-project / agentos

The Python Component System (PCS) is an API and CLI for building, running, and sharing Python code. AgentOS is a set of libraries built on top of PCS that make it easy to build, run, and share agents that use Reinforcement Learning.
https://agentos.org
Apache License 2.0
13 stars 4 forks source link

Create a hybrid agent that uses components from RL and cognitive architectures #116

Open andyk opened 3 years ago

andyk commented 3 years ago
  1. Create a design discussion with initial proposal
  2. Select target use case/env, benchmark
  3. Include components: Policy, Env, Trainer, MetaTrainer, RewardFunction, WorkingMemory, TaskController. It should be multi-task, hierarchical, associative, exploratory (intrinsically motivated), have both model free and model based subsystems.
andyk commented 3 years ago

@nickjalbert since you've started thinking about the benchmark/ usecase/env for the hybrid agent, I thought we might revisit a way to organize our work and make a specific plan that exists next to the current other two major prongs of AgentOS work (i.e., the repo design and core abstractions)